© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 4851
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ibal
    -0.28
    ira
    -0.27
    inally
    -0.26
    Äįe
    -0.26
    uais
    -0.25
    ÊĬ
    -0.24
    entlich
    -0.24
    æĺĤ
    -0.24
     Ep
    -0.23
    ommen
    -0.23
    POSITIVE LOGITS
    ABCDEFGHIJKLMNOP
    0.28
    罩
    0.25
    ç»ĵæŀĦæĢ§
    0.25
     Provid
    0.24
    ovid
    0.24
     Appe
    0.24
    ZY
    0.24
     ”↵↵
    0.23
    owell
    0.23
     gravid
    0.23
    Activations Density 0.002%

    No Known Activations